Multi-Sensor Data Fusion for Reliable UGV Operations in GNSS-Denied Areas

Abstract

The operation of Unmanned Ground Vehicles (UGVs) in Global Navigation Satellite System (GNSS)-denied areas presents significant challenges in navigation and situational awareness. This paper explores a multi-sensor data fusion framework that integrates information from various sensors, such as LIDAR, cameras, inertial measurement units (IMUs), and odometers, to enhance the reliability of UGV operations in challenging environments. By employing advanced algorithms for sensor fusion, we create a cohesive understanding of the vehicle's surroundings, facilitating precise localization and obstacle detection. The proposed framework uses a Bayesian approach to merge sensor data, enabling the UGV to maintain accurate positioning and navigate complex terrains without reliance on GNSS. Experimental results demonstrate the effectiveness of the multi-sensor fusion system in various scenarios, including urban environments, forested areas, and disaster-stricken regions, where GNSS signals may be weak or unavailable. This research provides valuable insights into improving the autonomy and robustness of UGVs in GNSS-denied conditions.